Virtual burst-switching networks

Abstract

A time-shared network comprising edge nodes and optical core nodes may be dynamically divided into several embedded networks, each of which covering selected edge nodes. At least one of the edge nodes may host an embedded-network controller operable to form multiple-source flow-rate allocation requests each of the requests specifying flow-rate allocations to a plurality of paths from several source nodes to several sink nodes. A core node may also host an embedded-network controller or several embedded-network controllers. The time-shared network may use both time-division multiplexing and burst switching.

Description

FIELD OF THE INVENTION[0001]The present invention relates to a data network and, in particular, to an agile burst-switching network that includes a multiplicity of embedded virtual burst-switching networks.BACKGROUND[0002]Prior-art optical burst switching has two main drawbacks: burst-transfer latency and burst loss. In a closed-loop scheme, a source node sends a request to a core node for transferring a burst, the request including a destination and size of the burst, and waits for a message from the core node, where the message acknowledges that the optical switch in the core node is properly configured, before sending the burst. In an open-loop scheme, the burst follows the burst transfer, request after a predetermined time period, presumably sufficient to schedule the burst transfer across the core, and it is expected that, when the burst arrives at the core node, the controller of the core node would have set up an internal path through the optical switch to a target output por...

AI Technical Summary

Benefits of technology

[0011]In a network of edge nodes interconnected through optical core nodes, time-division-multiplexing and burst transfer can coexist and both are enabled by time-locking each edge node to an adjacent core node or to a core node reached through channel switching. A fast-switching optical-core network having numerous edge nodes interconnected by optical core nodes may be divided into a plurality of embedded burst-switching “virtual networks”. Each virtual network is agile, having inter-nodal capacities which may be adapted to temporal and spatial traffic-load variations. To attain low delay and zero burst loss, the proximity of an edge node to the core node to which it is time-locked is used to determine whether a schedule for burst transfer from an edge node should be based on descriptors of waiting data bursts or anticipated data bursts having sizes determined according to prescribed flow rates.

Problems solved by technology

Prior-art optical burst switching has two main drawbacks: burst-transfer latency and burst loss.

The main drawback of this technique is the uncertainty of the fate of a burst thus transmitted.

Even if a very long time gap is kept between a burst-transfer request and the data burst itself, the lack of buffers at the core node may result in burst loss and a significant idle time.

Thus, in the closed-loop scheme, the time delay experienced in sending a burst transfer request and receiving an acceptance before sending a burst may be unacceptably high, leading to idle waiting periods and low network utilization in addition to requiring large storage at the edge nodes.

In the open-loop scheme, a burst may arrive at a core node before the optical switch can be configured to switch the burst and the burst may be lost.

Furthermore, the fact that the burst has been lost at the core node remains unknown to the source node for some time and a lost burst would have to be sent again after a predefined interval of time.

This renders closed-loop burst scheduling inappropriate.

The high round-trip delay requires that the source node have sizeable buffer storage.

On the other hand, open-loop burst scheduling, which overcomes the delay problem, can result in substantial burst loss due to unresolved contention at the core nodes.

Providing high-capacity wide-coverage time-shared networks using bufferless core nodes presents a significant challenge due to the difficulty of time-alignment at the bufferless core nodes.

Images